Epi-SSA: A novel epistasis detection method based on a multi-objective sparrow search algorithm

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Genome-wide association studies (GWAS) are crucial for understanding complex diseases.
• Current GWAS methods often focus on two-order epistatic interactions, limiting accuracy.
• High-order epistatic interactions play a significant role in disease development.

Purpose of the Study:

• To introduce Epi-SSA, a novel method for detecting high-order epistatic interactions.
• To evaluate Epi-SSA's performance against existing methods using simulation datasets.
• To apply Epi-SSA to real-world data for identifying disease-associated genes.

Main Methods:

• Epi-SSA utilizes a sparrow search algorithm optimized for multiple objective functions.
• The method was tested on five simulation datasets (DME 100, DNME 100, DME 1000, DNME 1000, DNME3 100) with varying complexity.
• Performance was compared against seven other established methods using F-measure.

Main Results:

• Epi-SSA achieved higher average F-measures (0.92-0.97) compared to existing methods (0.41-0.86) across all datasets.
• Epi-SSA's performance advantage increased with higher SNP numbers and epistasis order.
• Application to the WTCCC dataset identified potential disease-associated genes and gene pairs, some previously reported.

Conclusions:

• Epi-SSA is a potent and accurate tool for detecting high-order epistatic interactions.
• The method demonstrates superior performance in complex genetic analyses.
• Epi-SSA enhances the understanding of complex disease pathogenesis and gene interactions.